fix(inbox): break signal summary sections onto separate lines#2317
fix(inbox): break signal summary sections onto separate lines#2317MattPua wants to merge 1 commit into
Conversation
Insert line breaks around What's happening, Root cause, and How to resolve headers so report summaries are easier to scan. Co-authored-by: Cursor <cursoragent@cursor.com>
Prompt To Fix All With AIFix the following 2 code review issues. Work through them one at a time, proposing concise fixes.
---
### Issue 1 of 2
apps/code/src/renderer/features/inbox/components/utils/SignalReportSummaryMarkdown.tsx:27-32
The `listMarkdown` computation splits `raw` on the first newline, but after formatting, `raw` now begins with `**What's happening:**
…`, so `listMarkdown` becomes just `"**What's happening:**"`. The `list` variant then renders only the bold header (e.g. "What's happening:") with no actual summary content — a regression from the previous behaviour where the full inline text was visible. Using the pre-formatted content for the list split preserves the original single-line text for the list preview.
```suggestion
const rawContent = content?.trim() ? content : fallback;
const raw = formatSignalReportSummaryMarkdown(rawContent);
/** List rows: only the first line (before first newline); CSS still caps visual lines. */
const listMarkdown = rawContent.split(/\r?\n/)[0] ?? "";
```
### Issue 2 of 2
apps/code/src/renderer/features/inbox/utils/formatSignalReportSummaryMarkdown.test.ts:4-33
All four cases share the same `(input, expected)` shape and should be collapsed into a single `it.each` table, per the team's preference for parameterised tests.
Reviews (1): Last reviewed commit: "fix(inbox): break signal summary section..." | Re-trigger Greptile |
| const raw = formatSignalReportSummaryMarkdown( | ||
| content?.trim() ? content : fallback, | ||
| ); | ||
|
|
||
| /** List rows: only the first line (before first newline); CSS still caps visual lines. */ | ||
| const listMarkdown = raw.split(/\r?\n/)[0] ?? ""; |
There was a problem hiding this comment.
The
listMarkdown computation splits raw on the first newline, but after formatting, raw now begins with `What's happening:
…, so listMarkdownbecomes just"What's happening:". The list` variant then renders only the bold header (e.g. "What's happening:") with no actual summary content — a regression from the previous behaviour where the full inline text was visible. Using the pre-formatted content for the list split preserves the original single-line text for the list preview.
| const raw = formatSignalReportSummaryMarkdown( | |
| content?.trim() ? content : fallback, | |
| ); | |
| /** List rows: only the first line (before first newline); CSS still caps visual lines. */ | |
| const listMarkdown = raw.split(/\r?\n/)[0] ?? ""; | |
| const rawContent = content?.trim() ? content : fallback; | |
| const raw = formatSignalReportSummaryMarkdown(rawContent); | |
| /** List rows: only the first line (before first newline); CSS still caps visual lines. */ | |
| const listMarkdown = rawContent.split(/\r?\n/)[0] ?? ""; |
Prompt To Fix With AI
This is a comment left during a code review.
Path: apps/code/src/renderer/features/inbox/components/utils/SignalReportSummaryMarkdown.tsx
Line: 27-32
Comment:
The `listMarkdown` computation splits `raw` on the first newline, but after formatting, `raw` now begins with `**What's happening:**
…`, so `listMarkdown` becomes just `"**What's happening:**"`. The `list` variant then renders only the bold header (e.g. "What's happening:") with no actual summary content — a regression from the previous behaviour where the full inline text was visible. Using the pre-formatted content for the list split preserves the original single-line text for the list preview.
```suggestion
const rawContent = content?.trim() ? content : fallback;
const raw = formatSignalReportSummaryMarkdown(rawContent);
/** List rows: only the first line (before first newline); CSS still caps visual lines. */
const listMarkdown = rawContent.split(/\r?\n/)[0] ?? "";
```
How can I resolve this? If you propose a fix, please make it concise.| describe("formatSignalReportSummaryMarkdown", () => { | ||
| it("puts section body text on a new line after the header", () => { | ||
| const input = | ||
| "**What's happening:** Error tracking issue keyed on `app:dashboard_query`."; | ||
| expect(formatSignalReportSummaryMarkdown(input)).toBe( | ||
| "**What's happening:**\n\nError tracking issue keyed on `app:dashboard_query`.", | ||
| ); | ||
| }); | ||
|
|
||
| it("separates consecutive section headers onto their own lines", () => { | ||
| const input = | ||
| "**What's happening:** Users hit rate limits. **Root cause:** All four rate limiters are contended. **How to resolve:** Reduce blocking."; | ||
| expect(formatSignalReportSummaryMarkdown(input)).toBe( | ||
| "**What's happening:**\n\nUsers hit rate limits.\n\n**Root cause:**\n\nAll four rate limiters are contended.\n\n**How to resolve:**\n\nReduce blocking.", | ||
| ); | ||
| }); | ||
|
|
||
| it("separates a section header from preceding intro text", () => { | ||
| const input = | ||
| "Users on busy orgs are hitting hard limits. **What's happening:** Error tracking issue."; | ||
| expect(formatSignalReportSummaryMarkdown(input)).toBe( | ||
| "Users on busy orgs are hitting hard limits.\n\n**What's happening:**\n\nError tracking issue.", | ||
| ); | ||
| }); | ||
|
|
||
| it("leaves content without section headers unchanged", () => { | ||
| const input = "Plain summary with no structured sections."; | ||
| expect(formatSignalReportSummaryMarkdown(input)).toBe(input); | ||
| }); | ||
| }); |
There was a problem hiding this comment.
All four cases share the same
(input, expected) shape and should be collapsed into a single it.each table, per the team's preference for parameterised tests.
Context Used: Do not attempt to comment on incorrect alphabetica... (source)
Prompt To Fix With AI
This is a comment left during a code review.
Path: apps/code/src/renderer/features/inbox/utils/formatSignalReportSummaryMarkdown.test.ts
Line: 4-33
Comment:
All four cases share the same `(input, expected)` shape and should be collapsed into a single `it.each` table, per the team's preference for parameterised tests.
**Context Used:** Do not attempt to comment on incorrect alphabetica... ([source](https://app.greptile.com/review/custom-context?memory=instruction-0))
How can I resolve this? If you propose a fix, please make it concise.Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!
Problem
Signals had sections in the same line, which means it can be difficult to skim through quickly.
Changes
Insert line breaks around What's happening, Root cause, and How to resolve headers so report summaries are easier to scan.
How did you test this?
Locally.
Publish to changelog?